topic
stringclasses
2 values
relevance score
int64
1
10
paper name
stringlengths
19
239
text
stringlengths
1.56k
680k
synthetic_cpt
2
Targeted_Angular_Reversal_of_Weights_(TARS)_for_Knowledge_Removal_in_Large_Language_Models.pdf
4 2 0 2 c e D 6 1 ] L C . s c [ 2 v 7 5 2 0 1 . 2 1 4 2 : v i X r a Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models Harry J. Davies Department of Electrical Engineering Imperial College London London, UK harry.davies14@imperial.ac.uk Giorgos Iacovides Department of Electr...
synthetic_cpt
2
IdeaBench_Benchmarking_Large_Language_Models_for_Research_Idea_Generation.pdf
IdeaBench: Benchmarking Large Language Models for Research Idea Generation Sikun Guo*, Amir Hassan Shariatmadari*, Guangzhi Xiong, Albert Huang, Eric Xie, Stefan Bekiranov, Aidong Zhang University of Virginia {qkm6sq, ahs5ce, hhu4zu, kfa7fg, jrg4wx, sb3de, aidong}@virginia.edu 4 2 0 2 t c O 1 3 ] L C . s c [ 1 v ...
synthetic_cpt
1
GoldMiner_Elastic_Scaling_of_Training_Data_Pre-Processing_Pipelines_for_Deep_Learning.pdf
GOLD Mine: A new Galaxy Database on the WEB Gavazzi G., Boselli A., Donati A., Franzetti P., Scodeggio M. The galaxy database ”GOLDmine” (http://goldmine.mib.infn.it/) has been significantly up- dated (Sept/1/2003) (see ”Introducing GOLD Mine: A new Galaxy Database on the WEB” by Gavazzi et al. 2003, Astronomy & Astro...
synthetic_cpt
1
A_Closer_Look_at_Data_Augmentation_Strategies_for_Finetuning-Based_LowFew-Shot_Object_Detection.pdf
A Closer Look at Data Augmentation Strategies for Finetuning-Based Low/Few-Shot Object Detection Vladislav Li∗, Georgios Tsoumplekas†, Ilias Siniosoglou†‡, Vasileios Argyriou∗, Anastasios Lytos§, Eleftherios Fountoukidis§ and Panagiotis Sarigiannidis†‡ Abstract—Current methods for low- and few-shot object de- tectio...
synthetic_cpt
3
IterAlign_Iterative_Constitutional_Alignment_of_Large_Language_Models.pdf
4 2 0 2 r a M 7 2 ] L C . s c [ 1 v 1 4 3 8 1 . 3 0 4 2 : v i X r a ITERALIGN: Iterative Constitutional Alignment of Large Language Models Xiusi Chen1 Hongzhi Wen2 Sreyashi Nag3 Chen Luo3 Qingyu Yin3 Ruirui Li3 Zheng Li3 Wei Wang1 University of California, Los Angeles1 Michigan State University2 Amazon3 {xchen,w...
synthetic_cpt
4
CraftRTL_High-quality_Synthetic_Data_Generation_for_Verilog_Code_Models_with_Correct-by-Construction_Non-Textual_Representations_and_Targeted_Code_Repair.pdf
4 2 0 2 p e S 9 1 ] R A . s c [ 1 v 3 9 9 2 1 . 9 0 4 2 : v i X r a Preprint CRAFTRTL: HIGH-QUALITY SYNTHETIC DATA GENERATION FOR VERILOG CODE MODELS WITH CORRECT-BY-CONSTRUCTION NON-TEXTUAL REP- RESENTATIONS AND TARGETED CODE REPAIR Mingjie Liu∗, Yun-Da Tsai∗, Wenfei Zhou, Haoxing Ren NVIDIA Corporation {mingji...
synthetic_cpt
1
Evaluation_Metrics_of_Language_Generation_Models_for_Synthetic_Traffic_Generation_Tasks.pdf
Evaluation Metrics of Language Generation Models for Synthetic Traffic Generation Tasks Simone Filice2, Jason Ingyu Choi1, Giuseppe Castellucci1, Eugene Agichtein1, Oleg Rokhlenko1 1Amazon, Seattle - USA 2Amazon, Tel Aviv - Israel {filicesf,chojson,giusecas,eugeneag,olegro}@amazon.com 3 2 0 2 v o N 1 2 ] L C . s c ...
synthetic_cpt
1
Automated_Optical_Inspection_for_Printed_Circuit_Board_Assembly_Manufacturing_with_Transfer_Learning_and_Synthetic_Data_Generation.pdf
DVQI: A Multi-task, Hardware-integrated Artificial Intelligence System for Automated Visual Inspection in Electronics Manufacturing Audrey G. Chung1, Francis Li1, Jeremy Ward1, Andrew Hryniowski1,2,3, Alexander Wong1,2,3 1DarwinAI Corp., Waterloo, ON 2Vision and Image Processing Research Group, University of Waterloo ...
synthetic_cpt
3
Beyond_Classification_Financial_Reasoning_in_State-of-the-Art_Language_Models.pdf
0 1 0 2 r a M 1 2 ] R S . h p - o r t s a [ 1 v 2 0 0 4 . 3 0 0 1 : v i X r a Spectral Classification; Old and Contemporary Sunetra Giridhar Indian Institute of Astrophysics, Bangalore 560034, India Summary. Beginning with a historical account of the spectral classification, its re- finement through additional cr...
synthetic_cpt
3
Synthetic_Data_Augmentation_Using_Large_Language_Models_(LLM)_A_Case-Study_of_the_Kamyr_Digester.pdf
4 2 0 2 y a M 3 1 ] R I . s c [ 1 v 7 6 7 7 0 . 5 0 4 2 : v i X r a Synthetic Test Collections for Retrieval Evaluation Hossein A. Rahmani University College London London, UK hossein.rahmani.22@ucl.ac.uk Nick Craswell Microsoft Bellevue, US nickcr@microsoft.com Emine Yilmaz University College London & Amazon ...
synthetic_cpt
2
Mitigating_Bias_in_Large_Language_Models_A_Multi-Task_Training_Approach_Using_BERT.pdf
3 2 0 2 c e D 3 2 ] L C . s c [ 1 v 1 8 1 5 1 . 2 1 3 2 : v i X r a Multilingual Bias Detection and Mitigation for Indian Languages Ankita Maity1, Anubhav Sharma1, Rudra Dhar1, Tushar Abhishek1,2, Manish Gupta1,2, and Vasudeva Varma1 1 IIIT Hyderabad, India 2 Microsoft, India Abstract. Lack of diverse perspective...
synthetic_cpt
5
Knowledge_Distillation_Using_Frontier_Open-source_LLMs_Generalizability_and_the_Role_of_Synthetic_Data.pdf
KNOWLEDGE DISTILLATION USING FRONTIER OPEN-SOURCE LLMS: GENERALIZABILITY AND THE ROLE OF SYNTHETIC DATA 4 2 0 2 t c O 4 2 ] G L . s c [ 1 v 8 8 5 8 1 . 0 1 4 2 : v i X r a Microsoft Anup Shirgaonkar∗†, Nikhil Pandey†, Nazmiye Ceren Abay, Tolga Aktas, Vijay Aski ABSTRACT Leading open-source large language models...
synthetic_cpt
2
Can_Language_Models_Enable_In-Context_Database.pdf
PHONEME LEVEL LANGUAGE MODELS FOR SEQUENCE BASED LOW RESOURCE ASR Siddharth Dalmia, Xinjian Li, Alan W Black and Florian Metze Language Technologies Institute, Carnegie Mellon University; Pittsburgh, PA; U.S.A. {sdalmia|xinjianl|awb|fmetze}@cs.cmu.edu 9 1 0 2 b e F 0 2 ] L C . s c [ 1 v 3 1 6 7 0 . 2 0 9 1 : v i ...
synthetic_cpt
1
Solving_Quantitative_Reasoning_Problems_with_Language_Models.pdf
2 2 0 2 l u J 1 ] L C . s c [ 2 v 8 5 8 4 1 . 6 0 2 2 : v i X r a Solving Quantitative Reasoning Problems with Language Models Aitor Lewkowycz∗, Anders Andreassen†, David Dohan†, Ethan Dyer†, Henryk Michalewski†, Vinay Ramasesh†, Ambrose Slone, Cem Anil, Imanol Schlag, Theo Gutman-Solo, Yuhuai Wu, Behnam Neyshabu...
synthetic_cpt
1
Log-likelihood_Ratio_for_Low-Density_Parity-Check_Codes_Under_Binary_Symmetric_Erasure_Channel_in_DNA_Storage.pdf
1 1 0 2 p e S 8 1 ] S D . s c [ 1 v 0 9 8 3 . 9 0 1 1 : v i X r a A Dynamic Stabbing-Max Data Structure with Sub-Logarithmic Query Time Yakov Nekrich∗ Abstract In this paper we describe a dynamic data structure that answers one-dimensional stabbing- max queries in optimal O(log n/ log log n) time. Our data struc...
synthetic_cpt
1
PaLM_Scaling_Language_Modeling_with_Pathways.pdf
A Novel Remote Sensing Approach to Recognize and Monitor Red Palm Weevil in Date Palm Trees (manuscript) Yashu Kang1, Chunlei Chen1, Fujian Cheng1, Jianyong Zhang1 1 STAR VISION March 20, 2022 Abstract The spread of the Red Pal Weevil (RPW) has become an existential threat for palm trees around the world. I...
synthetic_cpt
1
Deep_Learning-based_Assessment_of_Oncologic_Outcomes_from_Natural_Language_Processing_of_Structured_Radiology_Reports.pdf
8 1 0 2 y a M 2 2 ] G L . s c [ 1 v 5 5 3 8 0 . 5 0 8 1 : v i X r a Opening the black box of deep learning Dian Lei , Xiaoxiao Chen , Jianfei Zhao School of Mechatronics Engineering and Automation, Shanghai University, Shanghai 200072, China Abstract The great success of deep learning shows that its technology ...
synthetic_cpt
2
ZMM-TTS_Zero-Shot_Multilingual_and_Multispeaker_Speech_Synthesis_Conditioned_on_Self-Supervised_Discrete_Speech_Representations.pdf
7 1 0 2 r p A 4 2 ] h p - p m o c . s c i s y h p [ 1 v 1 7 0 7 0 . 4 0 7 1 : v i X r a Performance Evaluation of the Zero-Multipole Summation Method in Modern Molecular Dynamics Software Shun Sakurabaa,∗, Ikuo Fukudab aGraduate School of Frontier Sciences, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa-shi,...
synthetic_cpt
2
CLIPSonic_Text-to-Audio_Synthesis_with_Unlabeled_Videos_and_Pretrained_Language-Vision_Models.pdf
CLIPSONIC: TEXT-TO-AUDIO SYNTHESIS WITH UNLABELED VIDEOS AND PRETRAINED LANGUAGE-VISION MODELS Hao-Wen Dong1,2∗ Xiaoyu Liu1 Jordi Pons1 Gautam Bhattacharya1 Santiago Pascual1 Joan Serr`a1 Taylor Berg-Kirkpatrick2 Julian McAuley2 1 Dolby Laboratories 2 University of California San Diego 3 2 0 2 l u J 3 2 ] D...
synthetic_cpt
3
Self-Evolved_Diverse_Data_Sampling_for_Efficient_Instruction_Tuning.pdf
1 0 0 2 r a M 9 2 1 v 5 4 2 3 0 1 0 / h t - p e h : v i X r a Non-abelian self-duality from self-interaction A. Khoudeir Instituto de F´ısica, Universidad Nacional Aut´onoma de M´exico Apdo. Postal 20-364, 01000 M´exico D. F. M´exico and Centro de Astrof´ısica Te´orica, Departamento de F´ısica, Facultad de Ciencia...
synthetic_cpt
8
Evaluating_Language_Models_as_Synthetic_Data_Generators.pdf
3 2 0 2 y a M 4 2 ] L C . s c [ 1 v 1 4 0 5 1 . 5 0 3 2 : v i X r a Generating Faithful Synthetic Data with Large Language Models: A Case Study in Computational Social Science Veniamin Veselovsky†, Manoel Horta Ribeiro†, Akhil Arora†, Martin Josifoski†, Ashton Anderson∗, Robert West† † EPFL ∗University of Toronto ...
synthetic_cpt
6
First_Train_to_Generate_then_Generate_to_Train_UnitedSynT5_for_Few-Shot_NLI.pdf
GenCo: Generative Co-training for Generative Adversarial Networks with Limited Data Kaiwen Cui*, Jiaxing Huang*, Zhipeng Luo, Gongjie Zhang, Fangneng Zhan, Shijian Lu† School of Computer Science Engineering, Nanyang Technological University {kaiwen001, zhipeng001}@e.ntu.edu.sg, {jiaxing.huang, Gongjiezhang, fnzhan, sh...
synthetic_cpt
2
Logic-LM_Empowering_Large_Language_Models_with_Symbolic_Solvers_for_Faithful_Logical_Reasoning.pdf
Abstraction Logic A New Foundation for (Computer) Mathematics Steven Obua obua@practal.com 2 2 0 2 l u J 2 1 ] O L . s c [ 1 v 0 1 6 5 0 . 7 0 2 2 : v i X r a Abstract. Abstraction logic is a new logic, serving as a foundation of mathematics. It combines features of both predicate logic and higher-order logic:...
synthetic_cpt
4
PULSAR_at_MEDIQA-Sum_2023_Large_Language_Models_Augmented_by_Synthetic_Dialogue_Convert_Patient_Dialogues_to_Medical_Records.pdf
PULSAR at MEDIQA-Sum 2023: Large Language Models Augmented by Synthetic Dialogue Convert Patient Dialogues to Medical Records Viktor Schlegel1,2, Hao Li2, Yuping Wu2, Anand Subramanian1,3, Thanh-Tung Nguyen1, Abhinav Ramesh Kashyap1, Daniel Beck4, Xiaojun Zeng2, Riza Theresa Batista-Navarro2, Stefan Winkler1,3 and Gor...
synthetic_cpt
2
The_Trade-offs_of_Domain_Adaptation_for_Neural_Language_Models.pdf
epl draft 6 1 0 2 c e D 5 ] R T . n i f - q [ 2 v 6 6 6 6 0 . 1 1 6 1 : v i X r a Quantifying immediate price impact of trades based on the k-shell decomposition of stock trading networks Wen-Jie Xie1,2,3, Ming-Xia Li2,3,4, Hai-Chuan Xu1,2,3, Wei Chen5, Wei-Xing Zhou1,3,6 (a) and H. Eugene Stanley7 1 Department ...
synthetic_cpt
3
Concept-skill_Transferability-based_Data_Selection_for_Large_Vision-Language_Models.pdf
Concept Generation in Language Evolution Martha Lewis, Jonathan Lawry Department of Engineering Mathematics, University of Bristol, BS8 1TR, UK martha.lewis@bristol.ac.uk, j.lawry@bristol.ac.uk 6 1 0 2 n a J 5 2 ] I A . s c [ 1 v 2 3 7 6 0 . 1 0 6 1 : v i X r a Abstract This thesis investigates the generation ...
synthetic_cpt
3
Filling_in_the_Gaps_LLM-Based_Structured_Data_Generation_from_Semi-Structured_Scientific_Data.pdf
Filling Gaps in Chaoti Time Series Fran es o Paparella Dipartimento di Matemati a (cid:16)E. de Giorgi(cid:17) Università di Le e Le e, Italy ∗ Abstra t We propose a method for (cid:28)lling arbitrarily wide gaps in deterministi time series. Cru ial to the method is the ability to apply Takens' theorem in o...
synthetic_cpt
2
Increasing_The_Performance_of_Cognitively_Inspired_Data-Efficient_Language_Models_via_Implicit_Structure_Building.pdf
Can humans help BERT gain “confidence”? Piyush Agrawal | 250944 Submitted for the degree of MSc Artificial Intelligence and Adaptive Systems University of Sussex 30th Aug, 2022 1 Declaration I hereby declare that this project has not been and will not be submitted in whole or in part to another ...
synthetic_cpt
1
Smoothie_Label_Free_Language_Model_Routing.pdf
SMOOTHIE: Label Free Language Model Routing Neel Guha∗1 Mayee F. Chen *1 Trevor Chow1 Ishan S. Khare1 Christopher Ré1 1Department of Computer Science, Stanford University December 9, 2024 Abstract 4 2 0 2 c e D 6 ] I A . s c [ 1 v 2 9 6 4 0 . 2 1 4 2 : v i X r a Large language models (LLMs) are increasingl...
synthetic_cpt
1
Effectiveness_of_Data_Augmentation_and_Pretraining_for_Improving_Neural_Headline_Generation_in_Low-Resource_Settings.pdf
Low-Resource Neural Headline Generation Ottokar Tilk and Tanel Alum¨ae Department of Software Science, School of Information Technologies, Tallinn University of Technology, Estonia ottokar.tilk@ttu.ee, tanel.alumae@ttu.ee 7 1 0 2 l u J 1 3 ] L C . s c [ 1 v 9 6 7 9 0 . 7 0 7 1 : v i X r a Abstract Recent neural...
synthetic_cpt
6
Aioli_A_Unified_Optimization_Framework_for_Language_Model_Data_Mixing.pdf
4 2 0 2 v o N 8 ] G L . s c [ 1 v 5 3 7 5 0 . 1 1 4 2 : v i X r a Aioli: A unified optimization framework for language model data mixing Mayee F. Chen*1 Michael Y. Hu⋆2 Nicholas Lourie3 Kyunghyun Cho2,3,4 Christopher Ré1 1Department of Computer Science, Stanford University 2Center for Data Science, New York ...
synthetic_cpt
1
Appeal_for_Attention_at_SemEval-2023_Task_3_Data_augmentation_extension_strategies_for_detection_of_online_news_persuasion_techniques.pdf
Neobility at SemEval-2017 Task 1: An Attention-based Sentence Similarity Model WenLi Zhuang ∗ Shan-Si Elementary School ChangHua County, Taiwan bibo9901@gmail.com Ernie Chang Department of Linguistics University of Washington Seattle, WA 98195, USA cyc025@uw.edu 7 1 0 2 r a M 6 1 ] L C . s c [ 1 v 5 6 4 5 0 . 3 ...
synthetic_cpt
3
SimRAG_Self-Improving_Retrieval-Augmented_Generation_for_Adapting_Large_Language_Models_to_Specialized_Domains.pdf
SimRAG: Self-Improving Retrieval-Augmented Generation for Adapting Large Language Models to Specialized Domains Ran Xu1,2*, Hui Liu2, Sreyashi Nag2, Zhenwei Dai2, Yaochen Xie2, Xianfeng Tang2, Chen Luo2, Yang Li2, Joyce C. Ho1, Carl Yang1, Qi He2 1 Emory University 2 Amazon {ran.xu,joyce.c.ho,j.carlyang}@emory.edu, li...
synthetic_cpt
4
Generating_Synthetic_Resume_Data_with_Large_Language_Models_for_Enhanced_Job_Description_Classification.pdf
Distilling Large Language Models using Skill-Occupation Graph Context for HR-Related Tasks Pouya Pezeshkpour1 Hayate Iso1 Thom Lake2 Nikita Bhutani1 Estevam Hruschka1 2Indeed {pouya,hayate,nikita,estevam}@megagon.ai 1Megagon Labs tlake@indeed.com 3 2 0 2 v o N 0 1 ] L C . s c [ 1 v 3 8 3 6 0 . 1 1 3 2 : v i X r...
synthetic_cpt
1
Automated_Depth_Dataset_Generation_with_Integrated_Quality_Metrics_for_Robotic_Manipulation.pdf
IEEE ROBOTICS AND AUTOMATION LETTERS. PREPRINT VERSION. ACCEPTED JUNE, 2022 1 TransCG: A Large-Scale Real-World Dataset for Transparent Object Depth Completion and A Grasping Baseline Hongjie Fang1, Hao-Shu Fang1, Sheng Xu1 and Cewu Lu2 2 2 0 2 g u A 8 2 ] O R . s c [ 2 v 1 7 4 8 0 . 2 0 2 2 : v i X r a Abstra...
synthetic_cpt
1
GLaM_Efficient_Scaling_of_Language_Models_with_Mixture-of-Experts.pdf
2 2 0 2 c e D 5 1 ] V C . s c [ 1 v 0 9 8 7 0 . 2 1 2 2 : v i X r a Full Contextual Attention for Multi-resolution Transformers in Semantic Segmentation Loic Themyr1,2 Clement Rambour1 Nicolas Thome1,3 Toby Collins2 Alexandre Hostettler2 1Conservatoire National des Arts et M´etiers, Paris, France 2IRCAD, Stras...
synthetic_cpt
5
Principle-Driven_Self-Alignment_of_Language_Models_from_Scratch_with_Minimal_Human_Supervision.pdf
4 9 9 1 y a M 0 2 ] O L . h t a m [ 1 v 4 0 2 5 0 4 9 / h t a m : v i X r a AN INDUCTION PRINCIPLE AND PIGEONHOLE PRINCIPLES FOR K-FINITE SETS Andreas Blass Abstract. We establish a course-of-values induction principle for K-finite sets in intuitionistic type theory. Using this principle, we prove a pigeonhole ...
synthetic_cpt
2
Training_language_models_to_follow_instructions_with_human_feedback.pdf
The Wisdom of Hindsight Makes Language Models Better Instruction Followers Tianjun Zhang * 1 Fangchen Liu * 1 Justin Wong 1 Pieter Abbeel 1 Joseph E. Gonzalez 1 Abstract Reinforcement learning has seen wide success in finetuning large language models to better align with instructions via human feedback. The so- calle...
synthetic_cpt
5
West-of-N_Synthetic_Preferences_for_Self-Improving_Reward_Models.pdf
The non-linear dual gravity equation of motion in eleven dimensions Keith Glennon and Peter West Department of Mathematics King’s College, London WC2R 2LS, UK Abstract We derive the non-linear dual graviton equation of motion in eleven dimensions in the context of E theory. E11 knows best. 0 2 0 2 n u J 3 ] h ...
synthetic_cpt
7
Real-Fake_Effective_Training_Data_Synthesis_Through_Distribution_Matching.pdf
5 1 0 2 t c O 8 ] T G . h t a m [ 3 v 8 2 9 5 . 2 0 2 1 : v i X r a Real open books and real contact structures Ferit ¨OZT ¨URK and Nermin SALEPC˙I Abstract. A real 3-manifold is a smooth 3-manifold together with an orientation preserving smooth involution, called a real struc- ture. In this article we study ope...
synthetic_cpt
1
FedSyn_Synthetic_Data_Generation_using_Federated_Learning.pdf
FedSyn: Synthetic Data Generation using Federated Learning Monik Raj Behera1, Sudhir Upadhyay1, Suresh Shetty1, Sudha Priyadarshini1, Palka Patel1, Ker Farn Lee1 {monik.r.behera,sudhir.x.upadhyay,suresh.shetty,sudha.priyadarshini} @jpmorgan.com 1Onyx by J.P. Morgan 2 2 0 2 r p A 6 ] L M . t a t s [ 2 v 1 3 9 5 ...
synthetic_cpt
2
Stay_on_topic_with_Classifier-Free_Guidance.pdf
COVID-19: Detecting Depression Signals during Stay-At-Home Period Jean Marie Tshimula,1 Belkacem Chikhaoui,1,2 Shengrui Wang1 1Department of Computer Science, Universit´e de Sherbrooke, QC J1K 2R1, Canada 2LICEF Research Center, Universit´e T ´ELUQ, QC H2S 3L5, Canada {kabj2801,shengrui.wang}@usherbrooke.ca belkacem.c...
synthetic_cpt
3
Self-Explained_Keywords_Empower_Large_Language_Models_for_Code_Generation.pdf
1 0 0 2 r a M 9 2 1 v 5 4 2 3 0 1 0 / h t - p e h : v i X r a Non-abelian self-duality from self-interaction A. Khoudeir Instituto de F´ısica, Universidad Nacional Aut´onoma de M´exico Apdo. Postal 20-364, 01000 M´exico D. F. M´exico and Centro de Astrof´ısica Te´orica, Departamento de F´ısica, Facultad de Ciencia...
synthetic_cpt
1
Synthesizing_Neural_Network_Controllers_with_Probabilistic_Model-Based_Reinforcement_Learning.pdf
Synthesizing Neural Network Controllers with Probabilistic Model-Based Reinforcement Learning Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek 8 1 0 2 g u A 1 ] O R . s c [ 3 v 1 9 2 2 0 . 3 0 8 1 : v i X r a Abstract— We present an algorithm for rapidly learning neural network policies for robotics sy...
synthetic_cpt
2
Scaling_Law_for_Post-training_after_Model_Pruning.pdf
P2 Law: Scaling Law for Post-Training After Model Pruning Xiaodong Chen1,2, Yuxuan Hu1,2, Xiaokang Zhang1,2, Yanling Wang3 Cuiping Li1,2, Hong Chen1,2, Jing Zhang1,2* 1School of Information, Renmin University of China, Beijing, China 2Key Laboratory of Data Engineering and Knowledge Engineering, Beijing, China 3 Zhong...
synthetic_cpt
5
Genie_Achieving_Human_Parity_in_Content-Grounded_Datasets_Generation.pdf
Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise Zhenghao Lin 1 2 Yeyun Gong 3 Yelong Shen 4 Tong Wu 5 2 Zhihao Fan 6 2 Chen Lin 1 Nan Duan 3 Weizhu Chen 4 3 2 0 2 b e F 7 1 ] L C . s c [ 2 v 5 8 6 1 1 . 2 1 2 2 : v i X r a Abstract In this paper, we intro...
synthetic_cpt
4
Differentially_Private_Knowledge_Distillation_via_Synthetic_Text_Generation.pdf
Differentially Private Knowledge Distillation via Synthetic Text Generation James Flemings Murali Annavaram University of Southern California {jamesf17, annavara}@usc.edu 4 2 0 2 n u J 5 ] G L . s c [ 2 v 2 3 9 0 0 . 3 0 4 2 : v i X r a Abstract Large Language models (LLMs) are achiev- ing state-of-the-art per...
synthetic_cpt
2
Emergence_of_In-Context_Reinforcement_Learning_from_Noise_Distillation.pdf
6 0 0 2 p e S 6 ] O A . n i l n [ 1 v 1 1 0 9 0 6 0 / n i l n : v i X r a Emergence is coupled to scope, not level Alex Ryan September 2006 Abstract Since its application to systems, emergence has been explained in terms of levels of observation. This approach has led to confusion, contradiction, incoherence a...
synthetic_cpt
3
Meta-Learning_the_Difference_Preparing_Large_Language_Models_for_Efficient_Adaptation.pdf
3 2 0 2 y a M 3 1 ] G L . s c [ 1 v 2 9 8 7 0 . 5 0 3 2 : v i X r a JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 1 DAC-MR: Data Augmentation Consistency Based Meta-Regularization for Meta-Learning Jun Shu, Xiang Yuan, Deyu Meng, and Zongben Xu Abstract—Meta learning recently has been heavily resea...
synthetic_cpt
4
Language_Models_can_Self-Lengthen_to_Generate_Long_Texts.pdf
4 2 0 2 c e D 9 ] L C . s c [ 2 v 2 7 8 8 1 . 6 0 4 2 : v i X r a EFFICACY OF LANGUAGE MODEL SELF-PLAY IN NON-ZERO-SUM GAMES Austen Liao∗, Nicholas Tomlin∗, & Dan Klein Computer Science Division University of California, Berkeley {austenliao,nicholas tomlin,klein}@berkeley.edu ABSTRACT Game-playing agents like Al...
synthetic_cpt
3
Effective_Large_Language_Model_Adaptation_for_Improved_Grounding_and_Citation_Generation.pdf
Effective Large Language Model Adaptation for Improved Grounding and Citation Generation Xi Ye♢∗ Ruoxi Sun♠ Sercan Ö. Arık♠ Tomas Pfister♠ ♢ The University of Texas at Austin ♠ Google Cloud AI ♢xiye@cs.utexas.edu ♠{ruoxis,soarik,tpfister}@google.com 4 2 0 2 r p A 2 ] L C . s c [ 3 v 3 3 5 9 0 . 1 1 3 2 : v i X ...
synthetic_cpt
1
MAUVE_Measuring_the_Gap_Between_Neural_Text_and_Human_Text_using_Divergence_Frontiers.pdf
1 2 0 2 v o N 3 2 ] L C . s c [ 3 v 4 5 4 1 0 . 2 0 1 2 : v i X r a MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontiers Krishna Pillutla1 Swabha Swayamdipta2 Rowan Zellers1 John Thickstun3 Sean Welleck1,2 Yejin Choi1,2 Zaid Harchaoui4 1Paul G. Allen School of Computer Science ...
synthetic_cpt
3
On_Extracting_Specialized_Code_Abilities_from_Large_Language_Models_A_Feasibility_Study.pdf
4 2 0 2 t c O 0 1 ] L C . s c [ 1 v 5 2 8 7 0 . 0 1 4 2 : v i X r a Preprint. EXTRACTING AND TRANSFERRING ABILITIES FOR BUILDING MULTI-LINGUAL ABILITY-ENHANCED LARGE LANGUAGE MODELS Zhipeng Chen1, Liang Song3, Kun Zhou2, Wayne Xin Zhao1∗, Bingning Wang3∗, Weipeng Chen3, Ji-Rong Wen1,2 1Gaoling School of Artificia...
synthetic_cpt
2
Language_Models_on_a_Diet_Cost-Efficient_Development_of_Encoders_for_Closely-Related_Languages_via_Additional_Pretraining.pdf
DIET: Lightweight Language Understanding for Dialogue Systems Tanja Bunk1∗ Daksh Varshneya1† Vladimir Vlasov1‡ Alan Nichol§ Rasa 0 2 0 2 y a M 1 1 ] L C . s c [ 3 v 6 3 9 9 0 . 4 0 0 2 : v i X r a Abstract Large-scale pre-trained language models have shown impressive results on language under- standing bench...
synthetic_cpt
1
AskIt_Unified_Programming_Interface_for_Programming_with_Large_Language_Models.pdf
AskIt: Unified Programming Interface for Programming with Large Language Models Katsumi Okuda CSAIL, MIT Cambridge, USA okuda@csail.mit.edu Mitsubishi Electric Corporation Amagasaki, Japan Saman Amarasinghe CSAIL, MIT Cambridge, USA saman@csail.mit.edu 3 2 0 2 c e D 7 2 ] L P . s c [ 2 v 5 4 6 5 1 . 8 0 3 2 : v i ...
synthetic_cpt
9
Synthetic_continued_pretraining.pdf
4 2 0 2 t c O 3 ] G L . s c [ 2 v 1 3 4 7 0 . 9 0 4 2 : v i X r a SYNTHETIC CONTINUED PRETRAINING Zitong Yang∗ Department of Statistics Stanford University Neil Band∗ Department of Computer Science Stanford University Shuangping Li Department of Statistics Stanford University Emmanuel Cand`es Department of Sta...
synthetic_cpt
1
Design_of_a_Neuronal_Training_Modeling_Language_Exemplified_with_the_AI-Based_Dynamic_GUI_Adaption.pdf
3 2 0 2 y a M 1 3 ] G L . s c [ 1 v 1 1 9 9 1 . 5 0 3 2 : v i X r a Neuron to Graph: Interpreting Language Model Neurons at Scale Alex Foote1∗, Neel Nanda2, Esben Kran1, Ioannis Konstas3, Shay B. Cohen4, Fazl Barez1,4,5∗ 1Apart Research 2Independent 3 School of Mathematical and Computer Sciences Heriot-Watt Unive...
synthetic_cpt
1
Training_data_augmentation_for_deep_learning_radio_frequency_systems.pdf
Data Augmentation for Deep Learning-based Radio Modulation Classification Liang Huang1, Weijian Pan1, You Zhang1, LiPing Qian1, Nan Gao2 and Yuan Wu3 9 1 0 2 c e D 0 1 ] P S . s s e e [ 2 v 6 2 0 3 0 . 2 1 9 1 : v i X r a Abstract— Deep learning has recently been applied to auto- matically classify the modulation ...
synthetic_cpt
3
Text2Traj2Text_Learning-by-Synthesis_Framework_for_Contextual_Captioning_of_Human_Movement_Trajectories.pdf
Text2Traj2Text: Learning-by-Synthesis Framework for Contextual Captioning of Human Movement Trajectories Hikaru Asano1,2* Ryo Yonetani3 1The University of Tokyo Taiki Sekii3 Hiroki Ouchi4,3,2 2RIKEN AIP 3CyberAgent Inc. 4Nara Institute of Science and Technology asano-hikaru19@g.ecc.u-tokyo.ac.jp, {yonetani_ryo, se...
synthetic_cpt
3
Phi-3_Technical_Report_A_Highly_Capable_Language_Model_Locally_on_Your_Phone.pdf
0 1 0 2 r a M 6 2 ] h p - o r t s a [ 4 v 2 8 6 4 . 1 1 7 0 : v i X r a Approximate wφ ∼ Ωφ Relations in Quintessence Models Mingxing Luo1∗ and Qi-Ping Su1,2† 1 Zhejiang Institute of Modern Physics, Department of Physics, Zhejiang University, Hangzhou, Zhejiang 310027, P R China 2 Key Laboratory of Frontiers in...
synthetic_cpt
2
Data_Quality_Control_in_Federated_Instruction-tuning_of_Large_Language_Models.pdf
4 2 0 2 t c O 5 1 ] G L . s c [ 1 v 0 4 5 1 1 . 0 1 4 2 : v i X r a Data Quality Control in Federated Instruction-tuning of Large Language Models Yaxin Du1 Rui Ye1 Fengting Yuchi1 Wanru Zhao2 Jingjing Qu3 Yanfeng Wang3,1 Siheng Chen1 ∗ 1 Shanghai Jiao Tong University 2 University of Cambridge 3 Shanghai A...
synthetic_cpt
1
Can_a_Large_Language_Model_Learn_Matrix_Functions_In_Context.pdf
Can a Large Language Model Learn Matrix Functions In Context? Paimon Goulart Computer Science and Engineering University of California Riverside Riverside, CA, USA paimon.goulart@email.ucr.edu Evangelos E. Papalexakis Computer Science and Engineering University of California Riverside Riverside, CA, USA epapalex@cs.u...
synthetic_cpt
1
PAITS_Pretraining_and_Augmentation_for_Irregularly-Sampled_Time_Series.pdf
9 9 9 1 r a M 2 ] C O . h t a m [ 1 v 2 2 0 3 0 9 9 / h t a m : v i X r a A Tuner that Accelerates Parameters Felipe M Pait1 and Paulo A Atkinson2 Universidade de S˜ao Paulo Laborat´orio de Automa¸c˜ao e Controle – pee S˜ao Paulo SP 05508–900 Brasil pait,atk@lac.usp.br Abstract ential equations (1) and (2), w...
synthetic_cpt
1
Advancing_Enterprise_Spatio-Temporal_Forecasting_Applications_Data_Mining_Meets_Instruction_Tuning_of_Language_Models_For_Multi-modal_Time_Series_Analysis_in_Low-Resource_Settings.pdf
3 2 0 2 y a M 5 ] M R . n i f - q [ 3 v 7 9 9 4 1 . 1 1 2 2 : v i X r a 1 A Comprehensive Survey on Enterprise Financial Risk Analysis from Big Data Perspective Yu Zhao*, Member, IEEE, Huaming Du*, Member, IEEE, Qing Li, Member, IEEE, Fuzhen Zhuang, Member, IEEE, Ji Liu, Member, IEEE, Gang Kou Abstract—Enterpri...
synthetic_cpt
1
Holdout-Based_Empirical_Assessment_of_Mixed-Type_Synthetic_Data.pdf
4 2 0 2 n u J 7 1 ] G L . s c [ 1 v 1 1 0 2 1 . 6 0 4 2 : v i X r a The Benefits and Risks of Transductive Approaches for AI Fairness Muhammed T. Razzak OATML University of Oxford Andreas Kirsch University of Oxford Yarin Gal OATML University of Oxford muhammed.razzak@cs.ox.ac.uk Abstract Recently, transduc...
synthetic_cpt
6
ALMA_Alignment_with_Minimal_Annotation.pdf
Mem. S.A.It. Vol. 75, 282 c(cid:13) SAIt 2008 Memorie della The exciting future of (sub-)millimeter interferometry: ALMA V. Casasola and J. Brand INAF – Istituto di Radioastronomia & Italian ALMA Regional Centre Via P. Gobetti 101, 40129 Bologna, Italy e-mail: casasola@ira.inaf.it; brand@ira.inaf.it interferometer...
synthetic_cpt
2
Low-Rank_Adaptation_with_Task-Relevant_Feature_Enhancement_for_Fine-tuning_Language_Models.pdf
9 9 9 1 l u J 0 3 1 v 9 6 5 7 0 9 9 / h p - p e h : v i X r a Spin dependent structure function g1 at low x and low Q2 B. Bade lek a,b J. Kiryluk b and J. Kwieci´nski c a Department of Physics, Uppsala University, P.O.Box 530, 751 21 Uppsala, Sweden b Institute of Experimental Physics, Warsaw University, Ho˙za 69...
synthetic_cpt
2
for_Going_Beyond_Nouns_With_Vision_&_Language_Models_Using_Synthetic_Data.pdf
3 2 0 2 g u A 0 3 ] V C . s c [ 2 v 0 9 5 7 1 . 3 0 3 2 : v i X r a Going Beyond Nouns With Vision & Language Models Using Synthetic Data Paola Cascante-Bonilla*†1,2 Khaled Shehada∗2,3 James Seale Smith2,4 Sivan Doveh6,7 Donghyun Kim2,7 Rameswar Panda2,7 G ¨ul Varol5 Aude Oliva2,3 Vicente Ordonez1 Rogerio Feri...
synthetic_cpt
2
On_the_Calibration_of_Large_Language_Models_and_Alignment.pdf
Does Alignment Tuning Really Break LLMs’ Internal Confidence? Hongseok Oh Wonseok Hwang* University of Seoul {cxv0519, wonseok.hwang}@uos.ac.kr 4 2 0 2 g u A 1 3 ] L C . s c [ 1 v 2 5 3 0 0 . 9 0 4 2 : v i X r a Abstract Large Language Models (LLMs) have shown remarkable progress, but their real-world ap- plicat...
synthetic_cpt
1
CLR-Bench_Evaluating_Large_Language_Models_in_College-level_Reasoning.pdf
8 1 0 2 r p A 8 2 ] G L . s c [ 1 v 2 4 7 0 1 . 4 0 8 1 : v i X r a Novel Prediction Techniques Based on Clusterwise Linear Regression Igor Gitman Machine Learning Department Carnegie Mellon University igitman@andrew.cmu.edu Eric Lei Machine Learning Department Carnegie Mellon University elei@andrew.cmu.edu Jie...
synthetic_cpt
3
SWITCH_Studying_with_Teacher_for_Knowledge_Distillation_of_Large_Language_Models.pdf
A linear state feedback switching rule for global stabilization of switched nonlinear systems about a nonequilibrium point Department of Mathematical Sciences, The University of Texas at Dallas 800 West Campbell Road Richardson, TX 75080 Oleg Makarenkov 8 1 0 2 n u J 2 2 ] C O . h t a m [ 1 v 4 4 8 8 0 . 6 0 8...
synthetic_cpt
1
Distributional_Learning_of_Variational_AutoEncoder_Application_to_Synthetic_Data_Generation.pdf
Learning Autoencoders with Relational Regularization Hongteng Xu * 1 2 Dixin Luo * 2 Ricardo Henao 2 Svati Shah 2 Lawrence Carin 2 0 2 0 2 n u J 6 2 ] G L . s c [ 4 v 3 1 9 2 0 . 2 0 0 2 : v i X r a Abstract A new algorithmic framework is proposed for learning autoencoders of data distributions. We minimize the...
synthetic_cpt
1
Optimizing_Handwritten_Digit_Recognition_with_CNN_and_Data_Augmentation_Strategies.pdf
Handwritten image augmentation Mahendran N IIT Tirupati mahendranNNM@gmail.com 3 2 0 2 g u A 6 2 ] V C . s c [ 1 v 1 9 7 3 1 . 8 0 3 2 : v i X r a Abstract In this paper, we introduce Handwritten augmentation, a new data augmentation for handwritten character images. This method focuses on augmenting handwritte...
synthetic_cpt
3
Prompting_to_Distill_Boosting_Data-Free_Knowledge_Distillation_via_Reinforced_Prompt.pdf
4 2 0 2 c e D 3 1 ] V C . s c [ 2 v 1 1 9 3 1 . 7 0 4 2 : v i X r a Continual Distillation Learning: An Empirical Study of Knowledge Distillation in Prompt-based Continual Learning Qifan Zhang, Yunhui Guo, Yu Xiang Department of Computer Science University of Texas at Dallas {qifan.zhang,yunhui.guo,yu.xiang}@utdal...
synthetic_cpt
5
Exploring_Mathematical_Extrapolation_of_Large_Language_Models_with_Synthetic_Data.pdf
Exploring Mathematical Extrapolation of Large Language Models with Synthetic Data Haolong Li* Tongji Universiy furlongli322@gmail.com Yu Ma Seed Foundation, ByteDance mayu.1231@bytedance.com Yinqi Zhang∗ East China Normal University zhang.inch@gmail.com Chen Ye† ESSC Lab, Tongji Universiy yechen@tongji.edu.cn Jie ...
synthetic_cpt
1
Style_Variation_as_a_Vantage_Point_for_Code-Switching.pdf
Style Variation as a Vantage Point for Code-Switching Khyathi Raghavi Chandu, Alan W Black Language Technologies Institute Carnegie Mellon University kchandu@cs.cmu.edu, awb@cs.cmu.edu 0 2 0 2 y a M 1 ] L C . s c [ 1 v 8 5 4 0 0 . 5 0 0 2 : v i X r a Abstract Code-Switching (CS) is a common phenomenon observed ...
synthetic_cpt
2
NB-MLM_Efficient_Domain_Adaptation_of_Masked_Language_Models_for_Sentiment_Analysis.pdf
Giant efficiency of long-range orbital torque in Co/Nb bilayers Fufu Liu1, Bokai Liang1, Jie Xu1, Chenglong Jia1,2†, Changjun Jiang1,2* 1 Key Laboratory for Magnetism and Magnetic Materials, Ministry of Education, Lanzhou University, Lanzhou 730000, China 2 Lanzhou Center for Theoretical Physi...
synthetic_cpt
7
Foundational_Large_Language_Models_for_Materials_Research.pdf
4 2 0 2 c e D 2 1 ] i c s - l r t m . t a m - d n o c [ 1 v 0 6 5 9 0 . 2 1 4 2 : v i X r a Foundational Large Language Models for Materials Research Vaibhav Mishra1,∗, Somaditya Singh1,∗, Dhruv Ahlawat1,∗, Mohd Zaki2,∗, Vaibhav Bihani3, Hargun Singh Grover3, Biswajit Mishra4, Santiago Miret5, Mausam1,3,#, N. M...
synthetic_cpt
2
Comparative_Study_on_Synthetic_and_Natural_Error_Analysis_with_BART_&_MarianMT.pdf
Grammatical vs Spelling error correction: An investigation into the responsiveness of Transformer based language models using BART and MarianMT Rohit Raju1,2, Peeta Basa Pati*,2, SA Gandheesh2, Gayatri Sanjana Sannala2 & Suriya KS2 1University of Colorado Boulder, CO, US, e-mail: rohit.raju@colorado.edu 2Departme...
synthetic_cpt
4
Diversify_and_Conquer_Diversity-Centric_Data_Selection_with_Iterative_Refinement.pdf
3 2 0 2 t c O 0 3 ] G L . s c [ 1 v 1 6 2 9 1 . 0 1 3 2 : v i X r a Diversify & Conquer: Outcome-directed Curriculum RL via Out-of-Distribution Disagreement Daesol Cho, Seungjae Lee, and H. Jin Kim Seoul National University Automation and Systems Research Institute (ASRI) Artificial Intelligence Institute of Seou...
synthetic_cpt
1
The_impact_of_altering_emission_data_precision_on_compression_efficiency_and_accuracy_of_simulations_of_the_community_multiscale_air_quality_model.pdf
Informatica 39 (2015) 501–505 501 A Multi-Agent based Approach for Simulating the Impact of Human Behaviours on Air Pollution Sabri Ghazi Laboratoire de gestion électronique du document, Computer Science department, University Badji Mokhtar, PO-Box 12, 23000,Annaba, Algeria E-mail: sabri.ghazi@univ-annaba.dz ...
synthetic_cpt
1
Density_Ratio_Estimation_via_Infinitesimal_Classification.pdf
Density Ratio Estimation via Infinitesimal Classification Kristy Choi˚ Chenlin Meng˚ Yang Song Stefano Ermon Computer Science Department, Stanford University 2 2 0 2 r a M 2 1 ] G L . s c [ 2 v 0 1 0 1 1 . 1 1 1 2 : v i X r a Abstract Density ratio estimation (DRE) is a funda- mental machine learning techniq...
synthetic_cpt
2
ReLM_Leveraging_Language_Models_for_Enhanced_Chemical_Reaction_Prediction.pdf
VALIDATING LARGE LANGUAGE MODELS WITH RELM 3 2 0 2 y a M 8 ] G L . s c [ 2 v 8 5 4 5 1 . 1 1 2 2 : v i X r a Michael Kuchnik 1 Virginia Smith 1 George Amvrosiadis 1 ABSTRACT Although large language models (LLMs) have been touted for their ability to generate natural-sounding text, there are growing concerns arou...
synthetic_cpt
2
NLPrompt_Noise-Label_Prompt_Learning_for_Vision-Language_Models.pdf
4 2 0 2 c e D 2 ] V C . s c [ 1 v 6 5 2 1 0 . 2 1 4 2 : v i X r a NLPrompt: Noise-Label Prompt Learning for Vision-Language Models Bikang Pan1,† Qun Li1,† Xiaoying Tang2 Wei Huang3 Zhen Fang4 Feng Liu5 Jingya Wang1 Jingyi Yu1 Ye Shi1,* 1ShanghaiTech University, Shanghai, China 2The Chinese University of Hong Ko...
synthetic_cpt
3
TabuLa_Harnessing_Language_Models_for_Tabular_Data_Synthesis.pdf
7 1 0 2 l u J 0 1 ] E S . s c [ 1 v 3 3 8 2 0 . 7 0 7 1 : v i X r a Tabula: A Language to Model Spreadsheet Tables Jorge Mendes and João Saraiva HASLab, INESC TEC and Universidade do Minho, Portugal {jorgemendes,saraiva}@di.uminho.pt Abstract. Spreadsheets provide a flexible and easy to use software de- velopme...
synthetic_cpt
1
Generation_of_Synthetic_Data_to_Improve_Security_Monitoring_for_Cyber-Physical_Production_Systems.pdf
0 2 0 2 v o N 7 2 ] Y S . s s i o e d e n o [ i t a c 1 i l b v u P 2 5 5 3 1 . 1 1 0 2 : v i X r a Design and Evaluation of A Cyber-Physical Resilient Power System Testbed Preprint, compiled November 30, 2020 Abhijeet Sahu1, Patrick Wlazlo2, Zeyu Mao1, Hao Huang1, Ana Goulart2, Katherine Davis1, and Saman Zonouz3...
synthetic_cpt
2
Going_Beyond_Nouns_With_Vision_&_Language_Models_Using_Synthetic_Data.pdf
3 2 0 2 g u A 0 3 ] V C . s c [ 2 v 0 9 5 7 1 . 3 0 3 2 : v i X r a Going Beyond Nouns With Vision & Language Models Using Synthetic Data Paola Cascante-Bonilla*†1,2 Khaled Shehada∗2,3 James Seale Smith2,4 Sivan Doveh6,7 Donghyun Kim2,7 Rameswar Panda2,7 G ¨ul Varol5 Aude Oliva2,3 Vicente Ordonez1 Rogerio Feri...
synthetic_cpt
3
Preference_Fine-Tuning_of_LLMs_Should_Leverage_Suboptimal_On-Policy_Data.pdf
Direct Preference Optimization with an Offset Afra Amini Tim Vieira Ryan Cotterell {afra.amini, ryan.cotterell}@inf.ethz.ch tim.f.vieira@gmail.com 4 2 0 2 n u J 6 ] L C . s c [ 2 v 1 7 5 0 1 . 2 0 4 2 : v i X r a Abstract Direct preference optimization (DPO) is a successful fine-tuning strategy for aligning...
synthetic_cpt
1
Auto-CORPus_Automated_and_Consistent_Outputs_from_Research_Publications.pdf
9 1 0 2 b e F 5 2 ] A Q . h t a m [ 1 v 8 9 4 9 0 . 2 0 9 1 : v i X r a SIMPLE CURRENT AUTO-EQUIVALENCES OF MODULAR TENSOR CATEGORIES CAIN EDIE-MICHELL Abstract. In this short note we investigate the process of constructing auto-equivalences of modular tensor categories using invertible objects. We derive condi...
synthetic_cpt
8
Unnatural_Instructions_Tuning_Language_Models_with_(Almost)_No_Human_Labor.pdf
Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor Or Honovichτ Thomas Scialomµ Omer Levyτ µ Timo Schickµ τ Tel Aviv University µ Meta AI 2 2 0 2 c e D 9 1 ] L C . s c [ 1 v 9 8 6 9 0 . 2 1 2 2 : v i X r a Abstract Instruction tuning enables pretrained language models to perform new t...
synthetic_cpt
7
Self-calibration_for_Language_Model_Quantization_and_Pruning.pdf
1 0 0 2 r a M 9 2 1 v 5 4 2 3 0 1 0 / h t - p e h : v i X r a Non-abelian self-duality from self-interaction A. Khoudeir Instituto de F´ısica, Universidad Nacional Aut´onoma de M´exico Apdo. Postal 20-364, 01000 M´exico D. F. M´exico and Centro de Astrof´ısica Te´orica, Departamento de F´ısica, Facultad de Ciencia...
synthetic_cpt
4
Data_Augmentation_for_Neural_Machine_Translation_using_Generative_Language_Model.pdf
Combining SMT and NMT Back-Translated Data for Efficient NMT Alberto Poncelas, Maja Popovi´c, Dimitar Shterionov, Gideon Maillette de Buy Wenniger and Andy Way School of Computing, DCU, ADAPT Centre {firstname.lastname}@adaptcentre.ie 9 1 0 2 p e S 9 ] L C . s c [ 1 v 0 5 7 3 0 . 9 0 9 1 : v i X r a Abstract Neur...
synthetic_cpt
2
BoostAdapter_Improving_Vision-Language_Test-Time_Adaptation_via_Regional_Bootstrapping.pdf
4 2 0 2 t c O 4 2 ] V C . s c [ 2 v 0 3 4 5 1 . 0 1 4 2 : v i X r a BoostAdapter: Improving Vision-Language Test-Time Adaptation via Regional Bootstrapping Taolin Zhang1 Jinpeng Wang 1 Hang Guo 1 Tao Dai∗ 2 Bin Chen 3 1 Tsinghua University 3 Harbin Institute of Technology Shu-tao Xia 1,4 2 Shenzhen University ...
synthetic_cpt
2
Test-Time_Alignment_via_Hypothesis_Reweighting.pdf
8 1 0 2 b e F 0 2 ] E S . s c [ 1 v 1 6 3 7 0 . 2 0 8 1 : v i X r a Fault Detection Effectiveness of Source Test Case Generation Strategies for Metamorphic Testing Prashanta Saha School of Computing, Montana State University Bozeman, Montana p66n633@msu.montana.edu Upulee Kanewala∗ School of Computing, Montana St...
synthetic_cpt
1
Grounding_Language_Models_to_Images_for_Multimodal_Generation.pdf
3 2 0 2 v o N 0 3 ] V C . s c [ 1 v 5 7 7 8 1 . 1 1 3 2 : v i X r a CoDi-2: In-Context, Interleaved, and Interactive Any-to-Any Generation Zineng Tang1,4* Ziyi Yang2† Mahmoud Khademi3 Yang Liu2 Chenguang Zhu3‡ Mohit Bansal4† 1UC Berkeley 2Microsoft Azure AI 3Zoom 4UNC Chapel Hill https://codi-2.github.io Abs...
synthetic_cpt
2
BoostAdapter_Improving_Training-free_Test-Time_Adaptation_via_Regional_Bootstrapping.pdf
4 2 0 2 t c O 4 2 ] V C . s c [ 2 v 0 3 4 5 1 . 0 1 4 2 : v i X r a BoostAdapter: Improving Vision-Language Test-Time Adaptation via Regional Bootstrapping Taolin Zhang1 Jinpeng Wang 1 Hang Guo 1 Tao Dai∗ 2 Bin Chen 3 1 Tsinghua University 3 Harbin Institute of Technology Shu-tao Xia 1,4 2 Shenzhen University ...
synthetic_cpt
2
Prompt_Discriminative_Language_Models_for_Domain_Adaptation.pdf
Domain-Agnostic Mutual Prompting for Unsupervised Domain Adaptation Zhekai Du1, Xinyao Li1, Fengling Li2, Ke Lu1, Lei Zhu3, Jingjing Li*1 1University of Electronic Science and Technology of China; 2University of Technology Sydney; 3Tongji University {zhekaid, xinyao326}@std.uestc.edu.cn, lijin117@yeah.net 4 2 0 2 r ...
synthetic_cpt
4
Training_Language_Models_on_Synthetic_Edit_Sequences_Improves_Code_Synthesis.pdf
4 2 0 2 t c O 5 1 ] G L . s c [ 2 v 9 4 7 2 0 . 0 1 4 2 : v i X r a TRAINING LANGUAGE MODELS ON SYNTHETIC EDIT SEQUENCES IMPROVES CODE SYNTHESIS Ulyana Piterbarg, Lerrel Pinto, Rob Fergus ∗ New York University up2021@cims.nyu.edu ABSTRACT Software engineers mainly write code by editing existing programs. In con...
synthetic_cpt
1
LLM4DS_Evaluating_Large_Language_Models_for_Data_Science_Code_Generation.pdf
LLM4DS: Evaluating Large Language Models for Data Science Code Generation Nathalia Nascimento EASER, Eng. Division Pennsylvania State University Great Valley, USA nqm5742@psu.edu Everton Guimaraes EASER, Eng. Division Pennsylvania State University Great Valley, USA ezt157@psu.edu Sai Sanjna Chintakunta EASER, Eng. D...
synthetic_cpt
2
P-Flow_A_Fast_and_Data-Efficient_Zero-Shot_TTS_through_Speech_Prompting.pdf
EPJ manuscript No. (will be inserted by the editor) 2 0 0 2 t c O 2 1 v 6 0 0 0 1 2 0 / x e - l c u n : v i X r a Decay Properties of the Roper Resonance from pp → ppπ+π− H. Clement representing the PROMICE/WASA collaboration Physikalisches Institut der Universit¨at T¨ubingen, Morgenstelle 14, D-72076 T¨ubingen ...